Employment and unemployment

Information on labour status (employment and unemployment) of Latvia population is acquired with the help of continuous Labour Force Survey (LFS). The main goal of the Labour Force Survey is to obtain information on the labour status of Latvia population – characterise the labour force by sex, age and level of educational, as well as to compile data on kinds of economic activities, occupations in the current (for employed) or last (for unemployed) place of work, and other indicators on labour market.

N/A

The LFS has two target populations:

1. Resident population of Latvia aged 15–89 (until 2020 (including) 15–74) that during the reference period are living in private households;
2. Private households, in which at least one member during the reference period is permanent resident of Latvia aged 15–89 (until 2020 (including) 15–74).

EU data on Eurostat website in section: Employment and unemployment

Data collection methods used in the Labour Force Survey:

• 1995–2005 – face-to-face interviews using paper questionnaires (Paper-and-Pencil Interviewing – PAPI);
• starting from 2006 – face-to-face interviews using portable computers (computer-assisted personal interviewing – CAPI);
• starting from 2007 – CAPI and telephone interviews (computer-assisted telephone interviewing – CATI);
• starting from 2018 – CAPI, CATI and online surveys (computer-assisted web interviewing – CAWI);
• as of 13 March 2020, with aim to limit spread of Covid-19 – CATI interviews and online surveys (CAWI).

The survey covers all persons living in the surveyed household, and questions on activity status are asked to persons aged 15–89 (prior to 2001, to persons aged 15 and over, but from 2002 until 2020 (including) – persons aged 15–74). The questionnaires contain relevant questions characterising the activity of the population; these questionnaires were prepared in accordance with the internationally approved methodology of the International Labour Organisation (ILO) specifically in the area of labour force surveys that ensures comparability of information with other countries. The LFS provides information on the number of population including the active population (employed and unemployed) broken down by various characteristics (sex, age, education qualification, place of residence, employment status, etc.).

Results of the Population and Housing Census 2011 testified that number of Latvia population is smaller than one used for the extrapolation of data acquired in the Labour Force Survey. Considering changes in the population number, taking place since the previous Population and Housing Census of 2000, and re-calculation of the population number in compliance with the results of the last Population and Housing Census, also data of the Labour Force Survey on years 2001–2011 is accordingly revised. Starting from the 1st quarter of 2012 results of the Labour Force Survey are published basing on the population number acquired in the Population and Housing Census 2011.

Data on 1996–2000 are published accordingly national approach – including in the calculations all persons surveyed and those serving in compulsory military service.

To achieve compliance with recommendations of Task Force on quality of the Labour Force Survey, extrapolation of the Labour Force Survey data is based only on number of persons living in private households, excluding collective households (old people’s homes, student hostels, hospitals, prisons etc.). Information on number of persons living in collective households was obtained within the framework of the Population and Housing Census and has been updated regulary.

Starting with 2014, for the Labour Force Survey quarterly data generalisation purposes, the quarterly average number of residents in private households is used (earlier – data at the beginning of the year). These methodological changes for the calculation of weights will in general not affect the time series.

Starting with 2021 for generalisation of quarterly data of the Labour Force Survey several changes have been carried out (see section “DATA COMPILATION”).

Monthly estimates

Unadjusted and seasonally adjusted monthly estimates on two groups of economic activity are published in the Official Statistics Portal (OSP) – on employed and unemployed aged 15–74 years.

Monthly indicators on employed and unemployed are available since 2002. Three indicators characterising economic activity by months and by sex are published in the Official Statistics Portal (OSP):

• number of unemployed;
• unemployment rate;
• number of employed.

Unadjusted and seasonally adjusted monthly indicators of the CSB on the number of employed and unemployed are elaborated basing on the International Labour Organization (ILO) methodology.

Monthly indicators of the CSB on the number of employed and unemployed are calculated indirectly using quarterly estimates of the Labour Force Sample Survey and monthly data on the registered unemployed provided by the State Employment Agency (SEA). To obtain unadjusted time series on indicators of the number of monthly employed and unemployed Chow–Lin (splitting-up) approach is used¹. Calculations are carried out applying R package tempdisagg.

¹ Chow, G.C. and Lin, A.-L. (1971) Best Linear Unbiased Interpolation, Distribution, and Extrapolation of Time Series by Related Series.

From 2002 to 2006 – 10 296 dwellings per year, from 2007 to 2012 – 24 128 dwellings per year, in 2013 – 26 676 dwellings per year, in 2014 – 29 588 dwellings per year, starting from 2015 to 2020 – 29 952 dwellings per year, from 2021 – 29 861 dwellings per year.

1995–2001

Sample

Labour Force Survey (LFS) was launched for the first time in November 1995¹. After that, LFS was organised twice a year – in May and in November.

The LFS sample was representative to residents of Latvia living in private households and aged 15–74 years.

A sample of dwellings was used for the LFS. In case of response, one household from each sampled dwelling and all corresponding household members aged 15 and over during reference period, took part in the survey.

The sample was formed as a rotating panel sample, where sampling units (dwellings) took part in the survey several times (three times in a row). The dwellings were replaced with other dwellings after the third time of interview, thus ensuring the rotation of dwellings in the panel.

Weighting

From 1995 to 2000, design weights were adjusted using the response homogeneity group method. Each primary sampling unit was used as a response homogeneity group for households.

Post-stratification method was used to obtain the final weights. In post-stratification method statistical data on permanent residents of Latvia at the beginning of the reference year in Riga, six cities and 26 administrative territory towns and rural areas in breakdown by age group and sex, were used as auxiliary information. Weight adjustment was performed at the personal level.

In 2001 design weights were adjusted using the response homogeneity group method. Each primary sampling unit was used as a response homogeneity group for households. Breakdown by stratum and survey time were defined as response homogeneity groups.

In 2001 in order to obtain final weights, calibration (ranking-ratio) method was applied. Weight adjustment was carried out at the household level, dividing a household into three age groups (0–14, 15–74, 74+). For calibration purposes statistical data on the number of population living in private households were used as auxiliary information taking into account Population and Housing Census 2011 results at the beginning of the period in breakdown by:

• 14 age groups;
• sex;
• territory type of place of residence (Riga, eight cities, towns and rural areas);
• region of the place of residence (Riga, Pierīga, Vidzeme, Latgale, Kurzeme and Zemgale).

The number of unemployed persons registered by the State Employment Agency in breakdown by sex and age group was used as auxiliary information for weight calibration.

Weight calibration method was implemented, using the statistical calculation environment R and set of procedures "Sampling".

2002–2006

Sample

In 2000 the CSB started elaboration of new LFS methodology in 2000². The new LFS methodology had the following key objectives:

1. to obtain quarterly estimates;
2. to organise the LFS as a continuous survey;
3. to implement efficient dispersion evaluators.

Based on the information on Population and Housing Census 2000 counting areas, a new sample design was developed, resulting in the creation of a frame of territories covering all private households of Latvia. These territories have been further on used as primary sample units. The territories were stratified into four strata: Riga, cities under state jurisdiction, towns and rural territories.

Two-stage sampling was used for the LFS. Territories were selected as primary sampling units at the first stage. Sampling of the territories was done with stratified systematic πps sampling, where elements are included in the sample with probabilities proportional to their size³. The second stage sampling units were dwellings. Dwellings were sampled with simple random sampling sample in each of the sampled territories.

In case of response from each dwelling included in the sample one household and members of the respective household who on Sunday of the reference week of the survey were 15–74 years old, took part in the LFS.

The main aim for using two-stage sampling was to reduce the survey costs. Design effect (which increased as the two-stage sample was used) was minimized with high number of primary sampling units in the sample.

The LFS was organised as a rotating panel survey. The rotation scheme of the households was similar to that before 2002. Dwellings took part in the survey three times with a 26-week interval.

Weighting

Design weights were adjusted using response homogeneity group method4. Each primary sampling unit was used as a response homogeneity group for households. Breakdown by stratum and survey time were defined as response homogeneity groups.

From 2002 to 2006 in order to obtain final weights, calibration (ranking-ratio) method was applied. Weight adjustment was carried out at the household level, dividing a household into three age groups (0–14, 15–74, 74+). For calibration purposes statistical data on the number of population living in private households were used as auxiliary information, taking into account Population and Housing Census 2011 results at the beginning of the period in breakdown by:

• 14 age groups;
• sex;
• territory type of place of residence (Riga, eight cities, towns and rural areas);
• region of the place of residence (Riga, Pierīga, Vidzeme, Latgale, Kurzeme and Zemgale).

The number of unemployed persons registered by the State Employment Agency in breakdown by sex and age group was used as auxiliary information for weight calibration.

Weight calibration method was implemented, using the statistical calculation environment R and set of procedures "Sampling". Each primary sampling unit (territory) was used as response homogeneity group for dwellings.

2007–2012

In order to obtain more precise estimates of the number of employed and unemployed persons in breakdown by region, sex and age group, as of 2007 the LFS sample size was increased 2.4 times. The rotation scheme was changed to ensure the overlap of the samples of successive quarters.

From 2007 onwards, dwellings participate in the LFS four times with a 13-week break, 39-week break and 13-week break. Such rotation scheme ensured the overlap of the samples of successive quarters and years.

During the time period between 2002 and 2009 the frame of territories used for the LFS sample was not updated. At the end of 2009 a study was preformed, which showed that the frame of territories was outdated. It did not characterise the target population of the LFS well enough due to population migration. A new frame of territories for sampling was developed at the end of 2009. The frame was developed using the previous frame of territories. The decision was made to perform stepwise update of the LFS sample in 2010.

The new sample of territories was sampled for the period 2010–2011. Dwellings participating in the LFS for the first time in 2010 were selected from the new frame of territories. Dwellings participating in the LFS for the first time before 2010 were selected from the old frame of territories. As a result two samples of territories were used for LFS 2010.

Before 2010 LFS sample was representative to residents of Latvia living in private households aged 15–74 years. The changes were made to the sampling frame starting from the fourth quarter of 2010 with an aim to make the LFS sample representative to all population of Latvia living in private households.

Since the LFS is a rotating panel survey, the period of five quarters (4th quarter of 2010 to 4th quarter of 2011) was needed to make the LFS sample representative to all population of Latvia living in private households. As of the first quarter of 2012 the LFS sample is representative to all population of Latvia living in private households.

Weighting

Design weights were adjusted using response homogeneity group method5. Each primary sampling unit was used as a response homogeneity group for households.Breakdown by stratum and survey time were defined as response homogeneity groups.

From 2007 to 2009 in order to obtain the final weights, the calibration (ranking-ratio) method was applied. Weight adjustment was carried out at the household level, dividing a household into three age groups (0–14, 15–74, 74+). For calibration purposes statistical data on the number of population living in private households were used as auxiliary information, taking into account Population and Housing Census 2011 results at the beginning of the period in breakdown by:

• 14 age groups;
• sex;
• territory type of place of residence (Riga, eight cities, towns and rural areas);
• region of the place of residence (Riga, Pierīga, Vidzeme, Latgale, Kurzeme and Zemgale).

The number of unemployed persons registered by the State Employment Agency in breakdown by sex and age group was used as auxiliary information for weight calibration.

Weight calibration method was implemented, using the statistical calculation environment R and set of procedures "Sampling".

As of the 1st quarter of 2010 weight adjustment was performed at the household level. It means that the weights for all persons of one household are equal. The weights of persons are equal to the corresponding household weight. The remaining weight formation methodology was not changed. In order to obtain the final weights, the calibration (ranking-ratio) method was applied. For calibration purposes statistical data on the number of population living in private households were used as auxiliary information, taking into account Population and Housing Census 2011 results at the beginning of the period in breakdown by:

• 14 age groups;
• sex;
• territory type of place of residence (Riga, eight cities, towns and rural areas);
• region of the place of residence (Riga, Pierīga, Vidzeme, Latgale, Kurzeme and Zemgale).

The number of unemployed persons registered by the State Employment Agency in breakdown by sex and age group was used as auxiliary information for weight calibration.

2013-2020

Sample

In order to ensure higher data quality for estimates providing a more detailed breakdown, during the time period from 2013 to 2014 the sample size was gradually increased, and in 2015 it 1.2 times exceeded the size of 2012.

During the time period from 2013 to 2015 the LFS sample and weight methodology has not been changed significantly. The LFS uses a rotating panel survey, which is representative to all population of Latvia living in private households. Dwellings participate in the LFS four times with a 13-week break, 39-week break and 13-week break.

Weighting

To obtain more precise statistical data the weight formation methodology has been improved.

In 2013 statistical data on population living in private households at the beginning of the reference year by the following breakdown were used as auxiliary information for weight calibration:

• 14 age groups;
• sex;
• territory type of place of residence (Riga, eight cities, towns and rural areas);
• region of the place of residence (Riga, Pierīga, Vidzeme, Latgale, Kurzeme and Zemgale) and eight cities (Daugavpils, Jelgava, Jēkabpils, Jūrmala, Liepāja, Rēzekne, Valmiera and Ventspils).

The number of unemployed persons registered by the State Employment Agency in breakdown by sex and age group was used as auxiliary information for weight calibration.

Starting from 2014

In turn, as of 2014, quarterly average statistical data on population living in private households preserving the above-mentioned breakdown by age, sex, territory type and region of place of residence has been used as auxiliary information for weight calibration.

As of the 3rd quarter of 2014 statistical data on the number of employed persons from the State Revenue Service database were added as auxiliary information for weight calibration, at the same time also using the registered number of unemployed persons provided by the State Employment Agency in breakdown by sex and age group.

Starting from 2015

In 2015, weight methodology was improved with additional non-response adjustment in breakdown by survey data collection form (direct interviews and telephone interviews).

Starting from 2017

As of 2017, the State Employment Agency data on people registered as unemployed persons are linked with the information acquired within the Labour Force Survey (thus reducing responding burden, since this information is not asked to respondents anymore). Thereinafter this information will be used in the weight calibration by linking it with the State Employment Agency data on the number of registered unemployed persons in breakdown by gender and age group. The rest of the methodology used in weight development is not changed and is comparable with the historical data.

Starting from 2021

Sample

During the time period from 2009 to 2019 list of territories was not renewed, from which territories for LFS sample were selected. In 2018 it was concluded that the list of territories is outdated – in the result of addressing arrangement process, demographic changes and new construction, it characterised LFS population researched insufficiently, thus in 2019 new list of territories for formation of samples was elaborated. New territories are formed in a shape of landfills, which have a defined shape, territory, border. This means that the list of dwellings, thus also of buildings, owned is known for the territory used in the sample. The new list of territories was tested during 2020 and a decision was taken to introduce it in the Labour Force Survey in 2021.

Dwellings that in 2021 participated in the LFS for the first time were selected using new list of territories. Dwellings that in 2021 participated in the survey, but for the first time participated in it before 2021, were saved in LFS. In the result, two samples of territories will be used in survey of 2021 and 1st quarter of 2022.

With 2021 LFS sample is formed independenly between years and it is negatively coordinated within the the year. Territories that are surveyed within the year, are selected once along with drawing sample for the first quarter of the year and are not overlapping during a year.

Dwellings take part four times in the LFS with interval of 13 weeks, 39 weeks and 13 weeks. A transition period of five quarters is needed (first quarter of 2021 – first quarter of 2022), so that all LFS sample would change to the sample of new territories. Starting with the second quarter of 2022, LFS sample is formed only from the new territories.

Weights

Taking into account changes of the legal regulation No. 2019/2240, which becomes effective as of 01.01.2021., and with an aim to acquire a more precise statistics, methodology of calculation of weights was improved. Several changes were made in the calculation of weights:

• Weights are calculated using registered place of residence of population (until now actual place of residence was used in the calculation). The change was made because starting with 2021 the CSB does not provide statistics on the resident population by their actual place of residence anymore. Territorial statistics of LFS still will be prepared by the actual place of residence of population.
• Ogre is classified as city under state jurisdiction (due to administrative-territorial reform, which became effective as of 1 July 2021). As well as the number of Ogre residents is also additionally used in weighting to improve accuracy on statistics of Ogre.
• The number of residents at a more detailed breakdown of age groups is used in weighting. Age group 0-14 is broken down in three age groups by five years (0-4, 5-9, 10-14). Age group 75 years and more are broken down in four age groups (75-79, 80-84, 85-89, 90 years and more).
• Information on the number of dwellings resided and size in 16 territories according to the information on the registered place of residence of population is used in weighting. This is done to improve accuracy for LFS statistics on private households. This is one of requirements of the new LFS implementation regulation.
• Number of employees and employers in age group 75-89 in breakdown by sex registered with SRS are used additionally for weighting.

Kind of preparation of weighting information has been changed. Weighting information is prepared on each week, then calculating average weekly indicators of the quarter. Such kind of preparation of weighting information is relevant to LFS methodology where population researched is observed with frequency in a week.

Wave approach

Sample

As of 2017, annual indicators in Labour Force Survey are acquired with the help of wave approach. Indicators the periodicity whereof in line with the Commission Implementing Regulation (EU) 2019/2240 (until 2020 – Commission Regulation (EC) No 377/2008) is year are acquired in the first interview only. Thus, in line with the sample design, in one quarter annual estimates are produced by asking information from one fourth of the quarterly sample.

Weights

In 2017, additional weights for indicators with annual periodicity are introduced.

In line with the sample design, initially quarterly design weights from the 1st wave respondents are created. Design weights are adjusted for each quarter by using response homogeneity group method.

The quarterly weights acquired are merged before calibration and divided by four. Calibration is used for annual data by using the raking-ratio method.

Statistics on usually resident population living in private households of Latvia in breakdown by five-year age groups and in territorial breakdown, as well as State Employment Agency (SEA) data on registered unemployed persons by age group is used as additional information in the weight calibration. To comply with the requirements of the Article 9 point 4(b), 5(b) of the Commission Implementing Regulation (EU) 2019/2240 (until 2020 - Article 3 in Annex I to the Regulation No 377/2008) regarding compliance of the totals, additionally also Labour Force Survey full-scope sample employment, unemployment and inactive population estimates in breakdown by gender and ten-year age group are used in weight calibration.